Insights about E-commerce - Review of Industry & Major Threats
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Insights about E-commerce Review of Industry & Major Threats
Table of Contents
01 E-commerce Industry Overview
1.1. E-commerce in Numbers
1.1.1. What is E-commerce
1.1.2. Market Size
1.1.3. Top Selling Industries
1.1.4. The Growth
1.2. Mobile Commerce
1.2.1. Types of M-commerce
1.2.2. The Boom
1.2.3. Undercover Secret of M-commerce
02 The Dark Side of E-commerce
2.1. Introduction to Payment Methods
2.2. Cybercrimes
2.3. How it Works
2.4. Merchants Liability to Fraud
2.5. Dimension of the Problem
2.6. How do Fraudsters Profile Merchants?
2.7. Why are Fraudsters Focusing on M-commerce?
203 Geographies
3.1. Statistics
3.2. Facts
04 How to Prevent Fraud in Online Payments
4.1. Payments are Changing
4.2. An Eye over European Payments
05 Our Solution
06 References
301 E-Commerce Industry
Overview
1.1. E-commerce in numbers
1.1.1. What is e-commerce
E-commerce mostly consists of electronic transactions related to the purchase and
delivery of goods and services. What can be considered an e-commerce practice:
Retailers with online presence – “e-tailers”;
Companies selling exclusively online;
Marketplace sellers like Amazon, eBay or Mercadolibre.
The most well-known form of e-commerce falls into the business to consumer (B2C)
category, which includes online retail or online shopping.
1.1.2. Market size
E-commerce is evidently not homogenous and for this reason, estimating the real
number of online merchants in the world is not a trivial mission. However, it is possible
to incorporate segmentation-decision processes to avoid misclassification of websites
based on the items they sell. According to Statista, in 2018:
There were 2-3 million 1.8 billion people Global e-retail
e-commerce companies in worldwide purchased sales amounted to
the world (excluding China) goods online $2.8 trillion
41.1.3. Top selling industries
Interestingly, fashion is the top selling 1 Clothing
industry in almost all countries with 2 Shoes
clothing1 being the highest contributing 3 Consumer electronics
product category. The increase in 4 Books, movies & music
5 Cosmetics & body care
demand in the fashion industry in the
6 Bags & accessories
online world owes itself mainly to the
7 Food & drinks
improvement in returns policies2. Following
8 Household appliances
clothing, shoes and consumer electronics
9 Furniture & household goods
are the 2nd and 3rd most popular categories,
10 Sports & outdoor
respectively.
Source: Statista
1.1.4. An on-growing industry
Consistent with eMarketer and Statista, in 2019 global e-commerce sales were expec-
ted to rise 20.7% to $3.5 trillion. As the chart shows, by 2021 this index will be landing
close to $5 trillion.
Asia-Pacific and Latin America will be the two regions with highest year-on-year
growth rates. Yet, Western Europe owns three of the top six e-commerce markets,
led by UK, Germany and France.
Retail e-commerce sales worldwide from 2014 to 2023 (in billion U.S. dollars)
7000 6542 *Forecast includes products or servi-
6000 5695 ces ordered using the internet or via
4927 any device, regardless the payment
5000
4206 method and excluding travel and
4000 3535
event tickets.
2982
Sales in billion U.S. dollars
3000
2382
Source: Statista
1845
2000 1548
1336
1000
0
2014 2015 2016 2017 2018 2019 2020* 2021* 2022* 2023*
1
Retrieved from Statista
2
Retrieved from Pipe Candy 51.2. Mobile commerce
Over time, mobile shopping has gained a tremendous share of the global e-commer-
ce market: customers are increasingly using their mobile devices for various online
shopping activities. The industries that are mostly affected by the growth of
m-commerce are financial services including mobile banking, telecommunication,
retail and information services.
1.2.1. Types of m-commerce
Mobile Shopping. Allows a customer to purchase from a
mobile device using a marketplace application such as
Amazon or through a web app.
Mobile Banking. Enables customers to conduct fanatical
transactions like bank transfers. Usually done though secure,
dedicated app provided by the financial institution.
Mobile Payments. Enables users to buy products in person
through digital wallets such as Apple Pay, Android Pay and
Samsung Pay.
1.2.2. The boom
Mobile retail commerce as percentage of retail e-commerce
63.5% 67.2% 70.4% 72.9%
2018 2019 2020 2021
In Q4 2018, desktop PCs accounted for approximately the same amount of global e-retail orders as smart-
phones. However, smartphones were the number one device in terms of retail website visits in 2018. This
can be explained by the intense demand during the Christmas holiday season.
Source: Statista
61.2.3. Undercover secret of m-commerce
There are various reasons that explain why this type of online commerce is growing
at such a fast pace.
1. Firstly, these devices enhance costumer flexibility: e-commerce already enables
access to a wider range of products and competitive prices; with mobile
commerce, these features are subject to an increased flexibility of spaces for
device usage. For this reason, m-commerce is sometimes referred to as
“contextual commerce” – listing products where customers are spending most
of their time.
2. On top of these bright benefits, there is also a variety of payment options, most
common being Apple Pay, PayPal One-Touch, Visa Checkout and Amazon Pay.
3. Ultimately, online merchants using the mobile channel have registered higher
conversion rates and ROI.
As long as they have a mobile device, they can shop whenever they want,
wherever they want.
702 The Dark Side of E-commerce
2.1. Introduction to payment methods
The vast diversity of payment methods is in fact one of the key drivers of the online
commerce growth. As so, these two industries go hand in hand and together they are
exposed to similar market risks.
The most common payment methods used in e-commerce are credit cards, mobile
payments, bank transfers and e-wallets, followed by those less popular such as
prepaid cards, direct deposit and cash4.
In m-commerce, the most used methods, are mobile wallets, contactless mobile
payments, closed loop mobile payments, money transfers, mobile point-of-sale and
carrier payments5.
2.2. Cybercrimes
Unfortunately, the consequences of the dimension and growth potential of
e-commerce are not only positive. Fraudsters can’t resist taking advantage, and
online payment fraud is rising fast. The most common cybercrimes are:
Card Not Present Fraud Account Takeover
Unauthorized use of payment card while Commonly known as “stolen identity”,
purchasing goods and services online or ATO occurs when a criminal gets a hold
via telephone. Stolen data includes on one’s information, such as usernames,
cardholder’s name, billing address, email addresses, passwords and social
account number, three digit security code security numbers. With this they are able
and the expiration date. According to to provide proof of their identity since they
Juniper Research, in Europe, 60% of card possess one’s personal details.
fraud are associated with CNP transactions.
4
Retrieved from PaymentWall: Payment Service Provider
5
Retrieved from BigCommerce: SaaS E-commerce Platform 82.3. How it works
Hacker steals
Fraudster uses
card/personal details
the stolen
to later sell/buy it on
information to
the dark web to/from
buy things online
fraudsters
1 2
Cardholder detects
Merchant validates
the fraudulent
transaction and
transaction and
ships the good to
contacts his/her
the fraudster
bank
3 4
Merchant
Cardholder’s
receives a charge-
bank contacts
back notification from
the merchant to
the cardholder’s
report the incident
bank
5 6
92.4. Merchant’s liability to fraud
Cyber payments lack the standard security measures that Card Present transactions
secure. Since this risk is higher, the merchant is liable for the acceptance of any
fraudulent order and incur into losses like:
Value of the product/service sold - Original fraudulent shipment;
Refund the scammed customer - Chargeback fee issued by the cardholder’s bank;
Penalty from acquiring bank - Merchant’s bank may charge a fee for every chargeback
received and increase this fee as the volume of fraudulent transactions goes up and
so does the number of chargebacks;
Customer experience - If by mistake the merchant rejects a legitimate customer
there is a high risk that the loss of such customer will not be only one-time, but rather
a loss of their lifetime shopping potential;
Manual reviews – Costs associated with time spent reviewing orders. Delaying an
order might decrease the chances of future purchases by that client.
2.5. Dimension of the problem
Retrospective Prospective
Merchants’ total fraud costs as a percentage of revenue went from 7.6% in 2016
to 8% in 20176.
In 2018, 82% of organizations reported problems related to payment fraud;
Large organizations were particularly vulnerable as business with over $1 billion
revenue registered a jump of 7% (from 80% to 87%)7.
In 2019, 42% of organizations experienced loss as a result of payment fraud;
There was a particular increase in fraud criminality in Automated Clearing House
(ACH) Fraud* via:
1. Debits (2018: 28% 2019: 33%)
2. Credits (2018: 13% 2019: 20%)7
6
Retrieved from Javelin Strategy & Research
7
Retrieved from Association for Financial Professionals Payments Fraud & Control Survey
*ACH Fraud – ACH is a tool used by financial institutions to handle direct deposits, checks or bill payments between businesses
10
and individuals; Through phishing, fraudsters obtain a checking account number and a bank routing number which they then use
to steal bank account passwords.Payment fraud is a persistent problem that is only getting worse despite
repeated warnings and educational outreach.
Fraud loss per card payment transaction in USA 2018 - by transaction
14 CP with PIN CP with signature or no CVM*
11,7
12
CNP* Overall
Fraud loss in U.S. cents per
10
8
In the United States, Card Not Present fraud is
6
4,6 the type of fraud with the biggest impact for the
4 merchants, which translates into a gross loss of
transaction
4,6
2 1,4 $11.7 cents per transaction in 2018.
0
Gross fraud loss
Retrospective Prospective
As statistics have been showing, the growth of fraud is and will continue to be
driven by the rapid expansion of the e-commerce market, higher money flows in
the online channel and the increased use of mobile payments.
112.6. How do fraudsters profile merchants?
Luxury goods & services Subscriptions
(e.g. possibility
of testing cards)
VODs & streaming
services
2.7. Why are fraudsters focusing on m-commerce?
Mobile devices, especially iPhones, present very similar characteristics - instal-
led software and plugins, graphical and screen settings - and are therefore
much harder to be distinguished from each other.
Mobiles are robust against device fingerprinting.
Old phones are cheaper and more difficult to track.
1203 Geographies
3.1. Statistics
Brasil UK USA
Total
211.4M 67.6M 329.1M
population
Internet
68% 87% 94%
penetration
Mobile
97% 72% 82%
penetration
Online
60M 48M 191M
shoppers
E-commerce
USD 15.3B USD 720.8B USD 547.7B
revenue
*All numbers as of 2018 and 2019
133.2. Facts
Brazil
Brazil is the largest e-commerce market in Latin America, ranked 10th in the world for
its sales. The most important players are B2W Digital, MercadoLivre and Alibaba. Pre-
ferred payment methods are credit cards and Boleto Bancário. The most common
types of fraud are directly related to credit cards – the levels are among the highest
in the world.
United Kingdom
The United Kingdom takes the lead of online shopping penetration in Europe and is
the third largest in the world. The leading multichannel retailers in 2019 are the
fashion retailer New Look, Schuh and Argos. MasterCard holds the largest share of
payments. The most common type of fraud is CNP. By 2021, 93% of the country’s
online users will purchase goods/services online.
USA
The world’s most developed market by far, the United States of America is the
global standard for e-commerce. The top American e-commerce platform are
Macy’s, Wish and Lowe’s. Visa is the preferred payment method. As e-commerce
and mobile commerce present impressive growth rates, the country sees more
than half of all global payment card fraud.
1404
How to Prevent Fraud
in Online Payments
4.1. Payments are changing
Traditionally, card payments usually required two steps:
Timeline Authorization – when cardholder’s bank approves the pay-
ment; Capture – when the card is charged.
Here’s how it works:
2001 Despite all efforts in security and verification systems, debit
and credit card payments had still a major exposure to fraud
risk. To tackle this problem, card networks implemented the
first version of what it’s known to be Authentication,
through a system called 3D Secure 1.
Customer is redirected
Customer inputs to a page where bank
3DS 1 card details to asks for a code of
confirm a payment password to approve
the purchase
Some famous examples
of branded 3DS 1 are Visa
Secure, Mastercard Identity
Check or American Express
SafeKey.
Advantages of 3DS1 for the Merchant:
Building an extra layer of fraud protection.
Ensuring that only legitimate users follow through the
pipeline.
15Shifting liability for chargebacks from merchants to the
customer’s bank.
Disadvantages of 3DS1 for the Merchant
Adding a step to the payment can add friction to the
checkout flow and lead customers to abandon the
purchase.
Some banks require from their cardholders to remember
passwords to complete authentication, which can be
easily forgotten and lead to higher rates of cart
abandonment.
4.2. An eye over European payments
In order to better protect consumers when they pay online,
October 8,
promote the use of innovative online and mobile payments
2015
and make cross-border European payment services safer,
the European Parliament adopted a legislation: The 2nd
Payment Services Directive (PSD2).
The most important regulatory requirement from this directive
is the need for Strong Customer Authentication (SCA) on
the majority of online payments, with the goal of reducing
fraud and make online payments more secure.
With SCA, a new form of authentication is employed – 3DS2
- which is mandatory whenever a payment isn’t eligible for
exemption* or when the bank denies an exemption request.
Like 3DS1, 3DS2 occurs at the very beginning of the transaction.
This time, the customer must respond to a two-factor
authentication prompt issued by their bank. This can be in
the form of:
* Payments such as fixed-amount subscriptions, phone sales, merchant-initiated transaction or general
low-risk transactions, can be exempt from SCA. 16Something Something Something
the customer the customer the customer
knows has is
(Password/PIN) (Hardware) (Fingerprint)
In order for the payment to go through, the transaction should
have at least two of these elements verified. Without it, many
payments may be declined by the bank of the merchant’s
client. Addionally, for authentication to happen, the customer
should be on the merchant’s website or mobile app.
Advantages of 3DS2 for the Merchant: “Frictionless
Authentication”
The merchant and their PSP are able to send more data
elements to the cardholder’s bank, easing the process of
making a decision about a transaction.
Automatic requests for eligible exemptions enable the
bank to assess a transaction before the prompt is sent
to the customer, avoiding friction and abandonment
chances.
September 14,
PSD2 went into effect!
2019
December 31,
Requirements are expected to be fully enforced.
2020
1705 Our Solution
At Nethone, we are passionate about helping merchants resolving their fraud
problems. We have developed proprietary AI-driven Know Your Users analytical
system that helps convert threats into well-informed and profitable decisions.
Our innovative cloud-based solution - Nethone Guard is based on 4 pillars:
Nethone Guard
Exhaustive user Seamless Thoroughly tested Human ingenuity
profiling data data hub Machine Learning at your disposal
Nethone Guard was designed to detect payment fraud by identifying anonymi-
sation, automation, and anomalous behaviour. Profiler is a proprietary solution that
gathers over 5,000 attributes and unique fingerprints about each user or device used
to access a website. This extensive set of attributes ranges from:
Hardware, Software and Browser Intelligence
E.g. GPU characteristics, GPU detection, virtual machine
detection, number of processor cores, mobile device
detection
18Behavioral Data
E.g. Mouse/touchpad movements, swipes, keystroke
dynamics, gyro readings, accelerometer, clipboard
usage
Network Intelligence
E.g. IP geolocation, TCP/IP stack analysis, connection
type detection (Wi-Fi, cellular), OS fingerprinting,
VPN/Proxy/Tunneling detection, Tor detection, public
IP leak
To analyse the data gathered through the whole user flow, we use advanced Machine
Learning (ML) algorithms. Attributes are analysed by our AI models to find non-obvious
links between them. Our profiler and models are carefully deployed and enriched by
the Data Scientist (DS) team within Nethone, working closely with our clients. They
assist whenever the merchant considers necessary and work on constant optimisation
of the models in order to make sure all key KPI of the client are met.
ML has proven to be more effective than other statistical rules with
proficiency to analyse highly unstructured data.
1906 References
https://searchmobilecomputing.techtarget.com/definition/m-commerce
https://www.jpmorgan.com/merchant-services/insights/re-
ports/european-overview
https://www.merchantriskcouncil.org/resource-center/whitepa-
pers/2017/the-future-of-digital-payments-in-europe
https://beeketing.com/blog/future-ecommerce-2019/
http://blog.paymentwall.com/guides/types-of-payment-me-
thods-for-ecommerce
https://www.bigcommerce.com/blog/mobile-commerce/#why-does-mo-
bile-commerce-matter
https://www.signifyd.com/resources/fraud-101/why-liable/
https://www.experian.com/assets/decision-analytics/re-
ports/global-fraud-report-2018.pdf
http://blog.pipecandy.com/e-commerce-companies-market-size/
https://www.emarketer.com/content/uk-ecommerce-2019
https://fesrvsd.fe.unl.pt:2058/topics/871/online-shopping/
https://www.prnewswire.com/news-releases/pay-
ments-fraud-jumps-to-record-high-82-of-businesses-impacted-survey-finds-
300825669.html
https://s3.amazonaws.com/dive_static/paychek/Financial_Im-
pact_of_Fraud_Study_FINAL.pdf
20https://www.thepaypers.com/
https://disfold.com/top-e-commerce-sites-us/
https://www.investopedia.com/terms/c/cardnotpresent-fraud.asp
https://www.onespan.com/blog/top-account-takeover-fraud-sche-
mes-and-how-protect-against-them
https://en.wikipedia.org/wiki/Payment_Services_Directive#Revi-
sed_Directive_on_Payment_Services_(PSD2)
https://stripe.com/en-pl/guides/sca-payment-flows
21You can also read
